A multi-electric aircraft high-average-peak ratio hybrid energy system hierarchical control method

Abstract

This invention discloses a hierarchical control method for a high peak-to-average power ratio (AMP) hybrid energy system in a multi-electric aircraft. For airborne hybrid energy systems, based on fully considering the real-time performance and priority of control objectives, different time scales are divided, and a hierarchical control method is used to optimize the overall economy, stability, and reliability of the system. In this invention, the hierarchical structure is as follows: the bottom layer is the physical layer, the middle layer is the improved droop control layer, and the top layer is the energy management layer. The physical layer performs local control over the bottom-level devices of the system. The improved droop control layer coordinates the distributed power sources and energy storage units of the physical layer to complete the control process. The energy management layer uses model predictive control to intelligently allocate the power of the hybrid energy system to each unit of the physical layer according to priority, completing the global optimization management decision of the system. This invention ensures the system's overall performance under high AMP dynamic loads and effectively handles the interference of uncertain factors on the control system, guaranteeing the overall operational performance of the system under hierarchical control.

Description

Technical Field

[0001] This invention belongs to the field of energy system operation control, and specifically relates to a hierarchical control method for a hybrid energy system with a high peak-to-average power ratio in multi-electric aircraft. Background Technology

[0002] As aircraft electrification increases, more-electric aircraft (MEAs) increasingly rely on electrical energy instead of hydraulic and pneumatic power as their secondary energy systems, necessitating high-capacity power supply systems. To achieve reliable 270V power, MEAs typically employ hybrid energy systems. Energy storage devices, characterized by high energy density, high power output, and rapid charging and discharging, are combined with onboard power electronics and distributed power sources such as generators to form a hybrid energy system. To achieve operational control and optimized design of the MEA hybrid energy system, the aircraft's power generation, distribution, and consumption are centralized within a unified system, planned, managed, and controlled by a controller. Therefore, the aircraft's power supply system can be considered an onboard microgrid. Hybrid energy systems enable efficient, rapid, and flexible application of distributed power sources. Their scalability and extensibility allow for the large-scale integration of a wide range of distributed power sources, achieving highly reliable supply to various types of loads and multiple energy forms, thus facilitating an effective transition to an active distribution network.

[0003] In recent years, hierarchical control has been first applied to DC microgrids by scholars both domestically and internationally. Generally, it consists of a system-level optimization layer, a decision-level control layer, and a physical layer. System-level optimization, combined with reliable energy management control strategies, ensures stable system operation over a large time scale. The goal of decision-level control is to maintain system power balance and DC bus voltage stability, achieving optimal overall system economy, stability, and reliability. Domestic and international experts have researched and developed different energy management strategies, which can be categorized into rule-based, optimization-based, and learning-based types based on different control methods. Optimization-based energy management strategies are more practical, as they have corresponding optimization objectives such as system efficiency and fuel economy. The control strategy that minimizes the objective function is calculated to optimize the hybrid power system. However, because airborne hybrid energy systems have a large number of high-power power electronic loads with strong load impact and high response time requirements, directly applying hierarchical control from microgrids to airborne hybrid energy systems makes it difficult for the control system to respond to high-amplitude, high-frequency load changes, resulting in large energy dissipation and reduced energy utilization. To address these issues, this invention proposes a hierarchical control structure for hybrid energy systems under high peak-to-average load ratios. By fully considering the demand response time and load priority of the control objectives, the system is divided into different time scales and matched with corresponding control levels, thereby facilitating the implementation of hierarchical control. Summary of the Invention

[0004] To address the issue of significant energy dissipation caused by untimely control system response when directly applying hierarchical control to hybrid energy systems, this invention extends hierarchical control to airborne hybrid energy systems and proposes a hierarchical control method for multi-electric aircraft hybrid energy systems under high peak-to-average load ratios. This method ensures the system operates under dynamic loads with high peak-to-average load ratios and effectively handles the interference of uncertain factors on the control system, guaranteeing the overall operational performance of the system under hierarchical control.

[0005] The technical solution of this invention is: a hierarchical control method for a high peak-to-average power ratio hybrid energy system in a multi-electric aircraft, comprising the following steps:

[0006] Step 1: Construct the physical layer of the airborne hybrid energy system, which is used for generator control, energy storage system charging and discharging control, and is responsible for transmitting data to the upper layer and receiving control from the upper layer;

[0007] Step 2: Construct a droop control layer, implement primary control through adaptive current control, and implement secondary control through secondary voltage compensation control;

[0008] Step 3: Build an energy management layer to achieve optimized allocation and management of the system.

[0009] Furthermore, in step 1, the physical layer of the airborne hybrid energy system includes a main generator (1), a main generator (2), an auxiliary generator, a controller (1), a controller (2), a controller (3), a controller (4), a controller (5), three sets of AC-DC converters, two sets of bidirectional DC-DC converters, a lithium-ion battery, a supercapacitor, and a load. The controllers (1), (2), and (3) control the AC-DC converters, and the controllers (4) and (5) control the bidirectional DC-DC converters. The main generator (1), the main generator (2), and the auxiliary generator are connected in parallel to the high-voltage DC bus through the three sets of AC-DC converters. The lithium-ion battery and the supercapacitor are connected in parallel to the high-voltage DC bus through the two sets of bidirectional DC-DC converters. The supercapacitor and the lithium-ion battery are charged by a three-stage generator. The load of the system is a dynamic load with a high peak-to-average power ratio.

[0010] Furthermore, in steps 1, 2, and 3, based on fully considering the real-time nature and priority of the control objectives, different time scales are divided. Within the millisecond time scale, a physical layer and an improved droop control layer are set up; within the second time scale, a model predictive control (MPC) algorithm is used as the control algorithm for the energy management layer.

[0011] Furthermore, in step 2, the mathematical model for adaptive current control is:

[0012]

[0013] Where: v oi For converter c i The actual output voltage; v * ref R is the bus reference voltage. * di Let λ be the initial value of the droop coefficient. i For the control variable of flow sharing control, the calculation formula is:

[0014]

[0015] Where: k pi and k ii For the current compensation control PI adjustment parameters, i oi For converter c i The output current value, σ oi is the current mismatch value; S is the complex domain.

[0016]

[0017] in: k is the average value of the output current. i This refers to the current distribution ratio.

[0018] Furthermore, in step 2, the mathematical model for the voltage secondary compensation control is as follows:

[0019]

[0020] Among them, v oi For converter c i Actual output voltage value; R di The actual value of the droop coefficient is calculated using the following formula:

[0021] R di =R * di -λ i

[0022] λ v For voltage control, the control variable is calculated as follows:

[0023]

[0024] Among them, v * ref The bus reference voltage; v o This represents the actual bus voltage; k pv k iv S represents the output voltage compensation PI regulation parameter; S is the complex domain.

[0025] Furthermore, in step 3, the energy management layer includes the following sub-steps:

[0026] Step 3.1: Set control, state, and output variables to establish a predictive model;

[0027]

[0028] In the formula, k is the current sampling time; Δu(k) is the change of the control quantity between the current time and the previous time; x(k+1) is the state variable matrix at the next time; y(k) is the output variable matrix at the current time; and A, B, and C are the state, input, and output matrices, respectively.

[0029] Δu(k)=u(k)-u(k-1)=[ΔP mg1 (k),ΔP mg2 (k),ΔP UC (k),ΔP B (k),ΔP ag (k)] T (10)

[0030] Where Δu(k) is the change in the control quantity between the current time and the previous time, and ΔP mg1 (k) represents the change in power of the main generator 1, ΔP mg2 (k) represents the change in power of the main generator 2, ΔP UC (k) represents the change in power of the supercapacitor module, ΔP B (k) represents the change in power of the lithium battery module, ΔP ag (k) represents the change in power of the auxiliary generator module;

[0031] Step 3.2: Set constraints, including system output power constraints, charge / discharge power constraints, and charge constraints;

[0032] Step 3.3: Set system output power constraints;

[0033] Step 3.4: Set the objective function. Based on the objective function, the optimal power allocation of each generator, lithium battery and supercapacitor in the system is finally obtained, realizing the intelligent optimal allocation of energy in the hybrid energy system of multi-electric aircraft.

[0034] Furthermore, in step 3.2, the system output power constraint is:

[0035] Assuming the system is lossless and the load power is two main generators (P) mg1 P mg2 ), auxiliary generator (P) ag ), lithium battery (P) B ) and supercapacitors (P UCThe sum of the power of ) satisfies:

[0036] P mg1 (k+t i |k)+P mg2 (k+t i |k)+P UC (k+t i |k)+P B (k+t i |k)+P ag (k+t i |k)=P load (11)

[0037] Among them, P mg1 (k+t i |k) represents the generator 1 at the current sampling time k to k+t. i Predicted power at time point, P mg2 (k+t i |k) represents the generator 2 at the current sampling time k to k+t. i Predicted power at time point, P UC (k+t i |k) represents the supercapacitor at the current sampling time k versus k+t. i Predicted power at time point, P B (k+t i |k) represents the lithium battery at the current sampling time k to k+t. i Predicted power at time point, P ag (k+t i |k) represents the auxiliary generator at the current sampling time k to k+t. i Predicted power at time point; P load This represents the load power.

[0038] Furthermore, in step 3.2, the charging and discharging power constraint is as follows:

[0039]

[0040] Among them, P mg1 (k+t i |k) represents the generator 1 at the current sampling time k to k+t. i Predicted power at time point, P mg2 (k+t i |k) represents the generator 2 at the current sampling time k to k+t. i Predicted power at time point, P UC (k+t i |k) represents the supercapacitor at the current sampling time k versus k+t. i Predicted power at time point, P B (k+ti |k) represents the lithium battery at the current sampling time k to k+t. i Predicted power at time point, P ag (k+t i |k) represents the auxiliary generator at the current sampling time k to k+t. i Predicted power at time point; P mg_MAX The maximum power of the main generator; P UC_MIN P UC_MAX These represent the minimum and maximum power of the supercapacitor, respectively; P B_MIN P B_MAX These are the minimum and maximum power of the lithium battery, respectively; P ag_MAX To supplement the generator power.

[0041] Furthermore, in step 3.2, the charge constraint is:

[0042]

[0043] SOC UC_MIN SOC UC_MAX The minimum and maximum states of charge (SOC) of a supercapacitor. B_MIN SOC B_MAX The minimum and maximum states of charge (SOC) of a lithium battery. UC (k+t i |k) represents the value of k+t at the current sampling time k. i Predicted state of charge (SOC) at time t. B (k+t i |k) represents the generator 2 at the current sampling time k to k+t. i Predicted value of state of charge at time 1

[0044] Furthermore, the objective function of step 3.4 is:

[0045]

[0046] In the formula, k = 0, 1, 2…; N is the prediction step size; Q is the positive definite weighting coefficient matrix of the prediction output error; P mean P represents the system's average power, which is also the system's reference trajectory. mg1 (k+t i |k)P mg2 (k+t i |k) represents the current sampling time k against k+t i The predicted output values ​​of the two main generators at time 1.

[0047] Invention Effects

[0048] The technical advantages of this invention are as follows: Based on fully considering the real-time nature and priority of the control objectives, different time scales are divided, and a hierarchical control method is used to optimize the overall economy, stability, and reliability of the system. In this invention, the hierarchical structure consists of a bottom layer (physical layer), a middle layer (improved droop control layer), and a top layer (energy management (EMS) layer). The physical layer performs local control over the bottom-level devices of the system. The improved droop control layer coordinates the control process of various distributed power sources and energy storage units within the physical layer. The energy management (EMS) layer uses model predictive control (MPC) to intelligently allocate power from the hybrid energy system to each unit in the physical layer according to priority, completing the global optimization management decision of the system. This invention ensures that the system can operate under dynamic loads with high peak-to-average power ratios and effectively handle the interference of uncertain factors on the control system, guaranteeing the overall operating performance of the system under hierarchical control. The load of the system is a dynamic load with a high peak-to-average power ratio, which places higher demands on the speed and stability of the system's power allocation. The time scales of the various layers are different. Within the millisecond time scale, the physical layer and improved droop control are set up, which has high requirements for real-time communication. Within the second time scale, the model predictive control (MPC) algorithm is used as the control algorithm for the energy management layer, which has low requirements for communication speed. Attached Figure Description

[0049] Figure 1 This is a diagram of a three-stage generator-lithium battery-supercapacitor architecture for an airborne hybrid energy system.

[0050] Figure 2 Power variation diagram for dynamic loads with high peak-to-average ratio.

[0051] Figure 3 This is a time-scale-based hierarchical control structure diagram.

[0052] Figure 4 The control block diagram for the improved droop control includes adaptive current control and voltage compensation control.

[0053] Figure 5 This is a diagram showing the droop coefficient adjustment characteristics of adaptive current control.

[0054] Figure 6 This is a diagram showing the droop coefficient adjustment characteristics of the voltage secondary compensation control.

[0055] Figure 7 For specific flight load profile diagram

[0056] Figure 8 Output power for each power supply unit in the physical layer of the hybrid energy system

[0057] Figure 9 Schematic diagram of DC bus voltage

[0058] Figure 10 Schematic diagram of lithium battery SOC

[0059] Figure 11 Schematic diagram of supercapacitor SOC Detailed Implementation

[0060] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0061] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0062] See Figures 1-11 This invention extends hierarchical control to airborne hybrid energy systems, proposing a hierarchical control method for multi-electric aircraft hybrid energy systems under high peak-to-average load ratios. This method ensures the system's performance under dynamic loads with high peak-to-average load ratios and effectively handles the interference of uncertain factors on the control system, thus guaranteeing the overall operational performance of the system under hierarchical control.

[0063] The layers are constructed as follows:

[0064] S1 sets the bottom layer as the physical layer, which specifically includes: a bidirectional DC-DC converter, an AC-DC converter, a lithium battery pack, a supercapacitor pack, a generator, and dynamic loads. The physical layer provides local control of the system's lower-level devices, such as generator control and energy storage system charging and discharging control, and is responsible for transmitting data to the upper layer and executing control commands issued by the upper layer.

[0065] S2 sets the intermediate layer as an improved droop control layer, which is divided into a first-level control aimed at accurately distributing the power of each unit in the system and a second-level control aimed at maintaining the bus voltage level. The two levels are the adaptive droop coefficient control layer and the voltage secondary compensation layer, respectively. Its main task is to coordinate the various distributed power sources and energy storage units in the physical layer to complete a series of basic controls.

[0066] S3 sets the top layer as the Energy Management System (EMS) layer, with a specific three-level control. Based on dynamic load demand, the operating status and characteristics of the two energy storage devices, and comprehensively considering control objectives and constraints, it realizes intelligent power allocation of the hybrid energy system according to priority, and completes the global optimization management decision of the system.

[0067] Furthermore, the load carried by the system is a dynamic load with a high peak-to-average power ratio, which places higher demands on the speed and stability of the system's power distribution.

[0068] Furthermore, the time scales of each layer are different. Within the millisecond time scale, the physical layer and the improved droop control layer are set up, which have high requirements for real-time communication. Within the second time scale, the model predictive control (MPC) algorithm is used as the control algorithm for the energy management layer, which has low requirements for communication speed.

[0069] Furthermore, the hierarchical control method applied to the airborne hybrid energy system specifically includes the following steps:

[0070] S1 builds an airborne hybrid energy system

[0071] The S11 hybrid energy system mainly consists of a three-stage brushless synchronous generator, a lithium-ion battery pack, a supercapacitor pack, an AC-DC converter, and a two-phase interleaved parallel bidirectional DC-DC converter. The three-stage generator, supercapacitor, and lithium-ion battery are connected in parallel to the high-voltage DC bus via the AC-DC converter and the bidirectional DC-DC converter, respectively. The supercapacitor and lithium-ion battery are charged by the three-stage generator.

[0072] The S12 system carries dynamic loads with high average-to-peak ratios, such as high-power radars and directed-energy weapons. High average-to-peak ratio loads place higher demands on the speed and stability of system power distribution.

[0073] S2 hierarchical control underlying construction

[0074] The bottom layer of the S21 hybrid energy system's hierarchical control structure is the physical layer, which consists of a bidirectional DC-DC converter, a lithium battery pack, a supercapacitor pack, a generator, and dynamic loads.

[0075] S22, the physical layer, provides local control over the underlying devices of the system, such as generator control and energy storage system charging and discharging control, and is responsible for transmitting data to the upper layer and receiving control from the upper layer.

[0076] S3 hierarchical control intermediate layer construction

[0077] The middle layer of the S31 hybrid energy system's hierarchical control structure is an improved droop control layer, comprising two levels: primary control and secondary control. Its main task is to coordinate the distributed power sources and energy storage units in the physical layer to complete a series of basic controls. The control strategy requires a short time scale, enabling real-time power balance of the system.

[0078] The main control objective of layer S32 is to maintain the stability of the DC bus voltage and achieve system power balance, thereby ensuring the stable operation of the hybrid energy system.

[0079] The improved droop control adds two layers of control to the traditional droop control, including adaptive current control and voltage compensation control. The control block diagram is shown in the figure.

[0080] The key input, output, and control variables involved in the control process are shown in the table below.

[0081] Table 1 Key inputs, outputs, and control variables in improved droop control.

[0082]

[0083] S33 Level 1 Control Design

[0084] The S331 Level 1 control is based on adaptive droop coefficient control. It collects the output current of each converter in the physical layer, calculates the average current value of each converter according to the current distribution ratio, and performs a PI calculation on the difference between this average current and the converter's output current. The result is used to adjust the droop coefficient R. di Multiple converters use C. i Let i = 1, 2, ..., n. The various variables in the control process are shown in the table.

[0085] The mathematical model for the current regulation control method of the S332 adaptive droop coefficient is as follows:

[0086]

[0087] Where: v oi For converter c i The actual output voltage; v * ref R is the bus reference voltage. * di Let λ be the initial value of the droop coefficient. i For the control variable of flow sharing control, the calculation formula is:

[0088]

[0089] Where: k pi and k ii For the current compensation control PI adjustment parameters, ioi For converter c i The output current value, σ oi is the current mismatch value; S is the complex domain.

[0090]

[0091] in: k is the average value of the output current. i This refers to the current distribution ratio.

[0092] Under the adaptive droop coefficient control, the S333 improves the power distribution accuracy, but the fluctuation of the DC bus voltage increases compared to the original. Based on this, a secondary control design is added.

[0093] S34 Level 2 Control Design

[0094] The S341 secondary control is an output voltage compensation control. It superimposes a voltage secondary compensation structure on the basis of the adaptive droop coefficient adjustment of the reference voltage in equation (1). The purpose is to make up for the voltage deviation caused by the primary control, thereby restoring the bus voltage level.

[0095] The mathematical model for the secondary compensation control of the S342 output voltage is as follows:

[0096]

[0097] Among them, v oi For converter c i Actual output voltage value; R di The actual value of the droop coefficient is calculated using the following formula:

[0098] R di =R * di -λ i

[0099] λ v For voltage control, the control variable is calculated as follows:

[0100]

[0101] Among them, v * ref The bus reference voltage; v o This represents the actual bus voltage; k pv k iv is the output voltage compensation PI regulation parameter; s is the complex domain.

[0102] S4 Hierarchical Control Top-Level Construction: Three-Level Control Design

[0103] The top layer of the hierarchical control structure of the hybrid energy system is the Energy Management System (EMS) layer, which has a three-level control system. Its main task is to allocate the power of the hybrid energy system according to priority based on dynamic load demand, the operating status and characteristics of the two energy storage devices, and comprehensive consideration of control objectives and constraints.

[0104] Level 3 control utilizes an online optimization-based control energy management strategy, specifically model predictive control (MMCC). MMCC acquires information about the control system through different predictive models and then achieves optimal control of the system through continuous online rolling optimization. This effectively handles the disturbances caused by uncertainties to the control system and exhibits strong adaptability to various operating conditions.

[0105] S41 sets control, state, and output variables to establish a predictive model.

[0106] S42 sets constraints

[0107] Based on the model predictive control strategy, considering the characteristics of the two energy storage devices, control and state constraints are set to provide optimized control signals for each module of the hybrid energy system.

[0108] S421 Set system output power constraint

[0109] S43 Set the objective function

[0110] Considering the safe and economical operation of the hybrid energy system, the system control objectives are mainly divided into two parts:

[0111] During system operation, S431 should maintain a balanced distribution of system power while meeting load requirements as much as possible at each sampling time, so as to ensure the normal operation of each unit.

[0112] To protect the generator's lifespan, S432 prioritizes maintaining a constant generator output power.

[0113] Ultimately, the optimal power allocation of the three generators, lithium batteries, and supercapacitors was obtained, realizing intelligent optimal energy allocation of the multi-electric aircraft hybrid energy system.

[0114] The technical content will be further explained below with specific examples.

[0115] Layered control of hybrid energy systems:

[0116] S1 constructs the physical layer of an airborne hybrid energy system.

[0117] The physical layer of the S11 hybrid energy system mainly consists of a three-stage brushless synchronous generator, a lithium-ion battery pack, a supercapacitor bank, an AC-DC converter, and a two-phase interleaved parallel bidirectional DC-DC converter. The system is as follows: Figure 1As shown, the three-stage generator, supercapacitor, and lithium-ion battery are connected in parallel to the high-voltage DC bus via an AC-DC converter and a bidirectional DC-DC converter, respectively. The supercapacitor and lithium-ion battery are charged by the three-stage generator.

[0118] The S12 system carries dynamic loads with high peak-to-average power ratios, such as high-power radars and directed-energy weapons. Specific load power is as follows: Figure 2 As shown, loads with high peak-to-average power ratios place higher demands on the speed and stability of system power distribution.

[0119] The S11 hybrid energy system mainly consists of a three-stage brushless synchronous generator, a lithium-ion battery pack, a supercapacitor bank, an AC-DC converter, and a two-phase interleaved parallel bidirectional DC-DC converter. The system is as follows: Figure 1 As shown, the three-stage generator, supercapacitor, and lithium-ion battery are connected in parallel to the high-voltage DC bus via an AC-DC converter and a bidirectional DC-DC converter, respectively. The supercapacitor and lithium-ion battery are charged by the three-stage generator.

[0120] The S12 system carries dynamic loads with high peak-to-average power ratios, such as high-power radars and directed-energy weapons. Specific load power is as follows: Figure 2 As shown, loads with high peak-to-average power ratios place higher demands on the speed and stability of system power distribution.

[0121] The S22 physical layer provides local control over the underlying devices of the system, such as generator control and energy storage system charging and discharging control. It is also responsible for transmitting physical layer hardware data to the middle layer and receiving control from the middle layer.

[0122] S3 hierarchical control intermediate layer construction

[0123] The middle layer of the S31 hybrid energy system's hierarchical control structure is an improved droop control layer, comprising two levels: primary control and secondary control. Its main task is to coordinate the distributed power sources and energy storage units in the physical layer to complete a series of basic controls. The control strategy requires a short time scale, enabling real-time power balance of the system.

[0124] The main control objective of layer S32 is to maintain the stability of the DC bus voltage and achieve system power balance, thereby ensuring the stable operation of the hybrid energy system.

[0125] The improved droop control adds a second layer of control to the traditional droop control, including adaptive current control and voltage compensation control. The control block diagram is shown below. Figure 4 As shown.

[0126] The key input, output, and control variables involved in the control process are shown in the table below.

[0127] Table 1 Key inputs, outputs, and control variables in improved droop control.

[0128]

[0129]

[0130] S33 Level 1 Control Design

[0131] The S331 Level 1 control is based on adaptive droop coefficient control. It collects the output current of each converter in the physical layer, calculates the average current value of each converter according to the current distribution ratio, and performs a PI calculation on the difference between this average current and the converter's output current. The result is used to adjust the droop coefficient R. di Multiple converters use C. i Let i = 1, 2, ..., n. The various variables in the control process are shown in Table 1.

[0132] The mathematical model for the current regulation control method of the S332 adaptive droop coefficient is as follows:

[0133]

[0134] Where: v oi For converter c i The actual output voltage; v * ref R* is the bus reference voltage. di Let λ be the initial value of the droop coefficient. i For the control variable of flow sharing control, the calculation formula is:

[0135]

[0136] Where: k pi and k ii For the current compensation control PI adjustment parameters, i oi For converter c i The output current value, σ oi is the current mismatch value; S is the complex domain.

[0137]

[0138] in: k is the average value of the output current. i This refers to the current distribution ratio.

[0139] Under the adaptive droop coefficient control, the S333 improves the power distribution accuracy, but the fluctuation of the DC bus voltage increases compared to the original. Based on this, a secondary control design is added.

[0140] Under the adaptive droop coefficient control of S333, the system improves power distribution accuracy, but the fluctuation of DC bus voltage increases compared to the original. Therefore, a secondary control design is added. For example... Figure 4 The figure shows the adaptive droop coefficient adjustment characteristic curve.

[0141] In the diagram: l1 represents converter c i Traditional droop control curve; R* di For converter c i The initial droop coefficient; I1 is the converter c i The initial output current; l2 is the converter c j Traditional droop control curve; R* dj For converter c j The initial droop coefficient; I2 is the converter c j Initial output current; l' 1,2 This is the droop control curve after adaptive droop adjustment; I' 1,2 For converter c i c j Output current after adaptive droop coefficient adjustment control.

[0142] S34 Level 2 Control Design

[0143] The S341 secondary control is an output voltage compensation control. It superimposes a voltage secondary compensation structure on the basis of the adaptive droop coefficient adjustment of the reference voltage in equation (1). The purpose is to make up for the voltage deviation caused by the primary control, thereby restoring the bus voltage level.

[0144] The mathematical model for the secondary compensation control of the S342 output voltage is as follows:

[0145]

[0146] Among them, v oi For converter c i Actual output voltage value; R di The actual value of the droop coefficient is calculated using the following formula:

[0147] R di =R * di -λ i

[0148] λ v For voltage control, the control variable is calculated as follows:

[0149]

[0150] Among them, v * ref The bus reference voltage; vo This represents the actual bus voltage; k pv k iv S represents the output voltage compensation PI regulation parameter; S is the complex domain.

[0151] The droop control regulation characteristics of S343 after voltage secondary compensation control are as follows: Figure 5 As shown.

[0152] In the diagram: curve l1 represents converter c. i Traditional droop control curve; R * di For converter c i The droop adjustment coefficient; R * dj For converter c j The droop adjustment coefficient; curve l2 represents the converter c j Traditional droop control curve; l'1 is the converter c i The droop control curve after output voltage compensation control; l'2 is the converter c j The droop control curve after adopting output voltage compensation control.

[0153] S4 Hierarchical Control Top-Level Construction: Three-Level Control Design

[0154] The top layer of the hierarchical control structure for a hybrid energy system is the Energy Management System (EMS) layer, which has three levels of control. Its main task is to allocate the power and state of charge of the hybrid energy system to the units in the physical layer according to priority, based on dynamic load demand, the operating status and characteristics of the two energy storage devices, and comprehensively considering control objectives and constraints. The variables with assignable values ​​are:

[0155] Table 2. Variables governed by the energy management layer.

[0156] Main generator 1 / 2 <![CDATA[Power value P mg1 、P mg2 <!-- 9 -->]]> Auxiliary generator <![CDATA[Power value P ag > Supercapacitor <![CDATA[Power value P UC , state of charge SOC UC > lithium batteries <![CDATA[Power value P B , state of charge SOC B >

[0157] The specific algorithm used in Level 3 control is Model Predictive Control (MPC). MPC strategies acquire information about the control system through different predictive models, and then achieve optimal control of the system through continuous online rolling optimization. This effectively handles the disturbances caused by uncertainties to the control system and has extremely strong adaptability to various operating conditions.

[0158] S41 sets control, state, and output variables to establish a predictive model.

[0159] At sampling time k, S411 takes the control variable u(k) as:

[0160] u(k)=[P mg1 (k),P mg2 (k),P UC (k),PB (k),P ag (k)] T (6)

[0161] Among them, P mg1 (k) represents the power of the main generator 1, P mg2 (k) represents the power of the main generator 2, P UC (k) represents the power of the supercapacitor module, P B (k) represents the power of the lithium battery module, P ag (k) represents the power of the auxiliary generator module.

[0162] The state variable matrix x(k) of S412 is:

[0163] x(k)=u(k)=[P mg1 (k), P mg2 (k),P UC (k),P B (k),P ag (k),SOC UC (k),SOC B (k)] T (7)

[0164] Among them, P mg1 (k) represents the power of the main generator 1, P mg2 (k) represents the power of the main generator 2, P UC (k) represents the power of the supercapacitor module, P B (k) represents the power of the lithium battery module, P ag (k) represents the power of the auxiliary generator module; SOC UC (k) represents the state of charge (SOC) of the supercapacitor. B (k) represents the state of charge of the lithium battery.

[0165] The output variable matrix y(k) of S413 is:

[0166] y(k)=[P mg1 (k)+P mg2 (k)+P UC (k)+P B (k)+P ag (k),P mg1 (k),P mg2 (k),SOC UC (k),SOC B (k)] T (8)

[0167] The discretized prediction model for the S414 hybrid energy system is as follows:

[0168]

[0169] In the formula, k is the current sampling time; Δu(k) is the change of the control quantity between the current time and the previous time; x(k+1) is the state variable matrix at the next time; y(k) is the output variable matrix at the current time; and A, B, and C are the state, input, and output matrices, respectively.

[0170] Δu(k)=u(k)-u(k-1)=[ΔP mg1 (k),ΔP mg2 (k),ΔP UC (k),ΔP B (k),ΔP ag (k)] T (10)

[0171] Where Δu(k) is the change in the control quantity between the current time and the previous time, and ΔP mg1 (k) represents the change in power of the main generator 1, ΔP mg2 (k) represents the change in power of the main generator 2, ΔP UC (k) represents the change in power of the supercapacitor module, ΔP B (k) represents the change in power of the lithium battery module, ΔP ag (k) represents the change in power of the auxiliary generator module.

[0172] S42 sets constraints

[0173] In optimization design, the objective function depends on the design variables, and the range of values ​​for these design variables is subject to various constraints, such as system power variables. Each constraint can be written as a function containing the design variables; these are called constraints.

[0174] Based on the model predictive control strategy, considering the characteristics of the two energy storage devices, control and state constraints are set to provide optimized control signals for each module of the hybrid energy system.

[0175] S421 sets system output power constraints

[0176] Assuming the system is lossless and the load power is two main generators (P) mg1 P mg2 ), auxiliary generator (P) ag ), lithium battery (P) B ) and supercapacitors (P UC The sum of the power of ) satisfies:

[0177] P mg1 (k+t i |k)+P mg2 (k+t i |k)+PUC (k+t i |k)+P B (k+t i |k)+P ag (k+t i |k)=P load (11)

[0178] Among them, P mg1 (k+t i |k) represents the generator 1 at the current sampling time k to k+t. i Predicted power at time point, P mg2 (k+t i |k) represents the generator 2 at the current sampling time k to k+t. i Predicted power at time point, P UC (k+t i |k) represents the supercapacitor at the current sampling time k versus k+t. i Predicted power at time point, P B (k+t i |k) represents the lithium battery at the current sampling time k to k+t. i Predicted power at time point, P ag (k+t i |k) represents the auxiliary generator at the current sampling time k to k+t. i Predicted power at time point; P load This represents the load power.

[0179] S422 sets charge / discharge power constraints:

[0180]

[0181] Among them, P mg1 (k+t i |k) represents the generator 1 at the current sampling time k to k+t. i Predicted power at time point, P mg2 (k+t i |k) represents the generator 2 at the current sampling time k to k+t. i Predicted power at time point, P UC (k+t i |k) represents the supercapacitor at the current sampling time k versus k+t. i Predicted power at time point, P B (k+t i |k) represents the lithium battery at the current sampling time k to k+t. i Predicted power at time point, P ag (k+t i |k) represents the auxiliary generator at the current sampling time k to k+t. iPredicted power at time point; P mg_MAX The maximum power of the main generator; P UC_MIN P UC_MAX These represent the minimum and maximum power of the supercapacitor, respectively; P B_MIN P B_MAX These are the minimum and maximum power of the lithium battery, respectively; P ag_MAX To supplement the generator power.

[0182] S423 sets charge state constraints:

[0183]

[0184] SOC UC_MIN SOC UC_MAX The minimum and maximum states of charge (SOC) of a supercapacitor. B_MIN SOC B_MAX The minimum and maximum states of charge (SOC) of a lithium battery. UC (k+t i |k) represents the value of k+t at the current sampling time k. i Predicted state of charge (SOC) at time t. B (k+t i |k) represents the generator 2 at the current sampling time k to k+t. i Predicted state of charge at time S43: Set objective function

[0185] Considering the safe and economical operation of the hybrid energy system, the system control objectives are mainly divided into two parts:

[0186] During system operation, S431 should maintain a balanced distribution of system power while meeting load requirements as much as possible at each sampling time, so as to ensure the normal operation of each distributed unit;

[0187] To protect the generator's lifespan, S432 prioritizes maintaining a constant generator output power.

[0188] The system's optimization model uses the difference between the controlled object's output value at future sampling points and the desired trajectory. Therefore, the optimization model satisfying the control objective, i.e., the objective function J, is defined as:

[0189]

[0190] In the formula, k = 0, 1, 2…; N is the prediction step size; Q is the positive definite weighting coefficient matrix of the prediction output error; P mean P represents the system's average power, which is also the system's reference trajectory. mg1 (k+t i |k)P mg2 (k+t i|k) represents the current sampling time k against k+t i The predicted output values ​​of the two main generators at time 1.

[0191] Ultimately, the optimal power allocation of the three generators, lithium batteries, and supercapacitors was obtained, realizing intelligent optimal energy allocation of the multi-electric aircraft hybrid energy system.

[0192] Specific Implementation Cases

[0193] To better verify the accuracy and feasibility of the control method, the invention is further described below with reference to the accompanying drawings and embodiments. By reading the references and consulting the specific parameters of the electrical equipment, a specific flight load profile is obtained as follows: Figure 7 As shown in Table 3, the specific parameters and system performance design of each physical layer unit are illustrated below:

[0194] Table 3 Specific parameters and design performance of each level

[0195]

[0196]

[0197] By using the data in the table and the hierarchical control methods at each level, and setting the objective function and constraints described in the specific implementation plan at the energy management layer, the power of each unit at different times under this flight profile of the aircraft was obtained, such as... Figure 8 The figure shows the output power of each power supply unit in the physical layer of the hybrid energy system. The load driven by the system is a dynamic load with a high peak-to-average power ratio. The total output power of the system meets the load requirements of a peak power of 400 kW for 30 seconds and a peak-to-average power ratio of 5:1. Furthermore, the output power of the three generators is stable, avoiding damage to the generator lifespan caused by large power fluctuations.

[0198] like Figure 9 As mentioned above, while meeting the dynamic load requirements at various times, the DC bus voltage of the system fluctuates between 269 and 272V. The droop control layer in the hierarchical control achieves system power balance while maintaining the stability of the DC bus voltage.

[0199] Among them, the SOC of lithium batteries and supercapacitors are as follows: Figure 10 and Figure 11 As shown.

[0200] The above description is merely a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A hierarchical control method for a high peak-to-average power ratio hybrid energy system in a multi-electric aircraft, characterized in that, Includes the following steps: Step 1: Construct the physical layer of the airborne hybrid energy system, which is used for generator control, energy storage system charging and discharging control, and is responsible for transmitting data to the upper layer and receiving control from the upper layer; Step 2: Construct a droop control layer, implementing primary control through adaptive current control and secondary control through secondary voltage compensation control; wherein: The mathematical model for adaptive current control is as follows: Where: v oi For converter c i The actual output voltage; v * ref R is the bus reference voltage. * di Let λ be the initial value of the droop coefficient. i For the control variable of flow sharing control, the calculation formula is: Where: k pi and k ii For the current compensation control PI adjustment parameters, i oi For converter c i The output current value, σ oi S represents the current mismatch value; S is the complex region. in: k is the average value of the output current. i For current distribution ratio; The mathematical model for voltage secondary compensation control is as follows: Among them, v oi For converter c i Actual output voltage value; R di The actual value of the droop coefficient is calculated using the following formula: R di =R * di -l i λ v For voltage control, the control variable is calculated as follows: Among them, v * ref The bus reference voltage; v o k represents the actual bus voltage. pv k iv The output voltage compensation PI adjustment parameter; S is the complex domain; Step 3: Build an energy management layer to achieve optimized allocation and management of the system.

2. The hierarchical control method for a high peak-to-average power ratio hybrid energy system of a multi-electric aircraft as described in claim 1, characterized in that, In step 1, the physical layer of the airborne hybrid energy system includes a main generator (1), a main generator (2), an auxiliary generator, a controller (1), a controller (2), a controller (3), a controller (4), a controller (5), three sets of AC-DC converters, two sets of bidirectional DC-DC converters, a lithium-ion battery, a supercapacitor, and a load. The controllers (1), (2), and (3) control the AC-DC converters, and the controllers (4) and (5) control the bidirectional DC-DC converters. The main generator (1), the main generator (2), and the auxiliary generator are connected in parallel to the high-voltage DC bus through the three sets of AC-DC converters. The lithium-ion battery and the supercapacitor are connected in parallel to the high-voltage DC bus through the two sets of bidirectional DC-DC converters. The supercapacitor and the lithium-ion battery are charged by a three-stage generator. The load of the system is a dynamic load with a high peak-to-average power ratio.

3. The hierarchical control method for a high peak-to-average power ratio hybrid energy system of a multi-electric aircraft as described in claim 1, characterized in that, In steps 1, 2, and 3, based on fully considering the real-time performance and priority of the control objectives, different time scales are divided. Within the millisecond time scale, a physical layer and an improved droop control layer are set up; within the second time scale, the model predictive control (MPC) algorithm is used as the control algorithm for the energy management layer.

4. The hierarchical control method for a high peak-to-average power ratio hybrid energy system of a multi-electric aircraft as described in claim 1, characterized in that, Step 3, the energy management layer includes the following sub-steps: Step 3.1: Set control, state, and output variables to establish a predictive model; In the formula, k is the current sampling time; Δu(k) is the change of the control quantity between the current time and the previous time; x(k+1) is the state variable matrix at the next time; y(k) is the output variable matrix at the current time; and A, B, and C are the state, input, and output matrices, respectively. Δu(k)=u(k)-u(k-1)=[ΔP mg1 (k),ΔP mg2 (k),ΔP UC (k),ΔP B (k),ΔP ag (k)] T (10) Where Δu(k) is the change in the control quantity between the current time and the previous time, and ΔP mg1 (k) represents the change in power of the main generator 1, ΔP mg2 (k) represents the change in power of the main generator 2, ΔP UC (k) represents the change in power of the supercapacitor module, ΔP B (k) represents the change in power of the lithium battery module, ΔP ag (k) represents the change in power of the auxiliary generator module; Step 3.2: Set constraints, including system output power constraints, charge / discharge power constraints, and charge constraints; Step 3.3: Set system output power constraints; Step 3.4: Set the objective function. Based on the objective function, the optimal power allocation of each generator, lithium battery and supercapacitor in the system is finally obtained, realizing the intelligent optimal allocation of energy in the hybrid energy system of multi-electric aircraft.

5. The hierarchical control method for a high peak-to-average power ratio hybrid energy system of a multi-electric aircraft as described in claim 1, characterized in that, In step 3.2, the system output power constraint is: Assuming the system is lossless and the load power is two main generators (P) mg1 P mg2 ), auxiliary generator (P) ag ), lithium battery (P) B ) and supercapacitors (P UC The sum of the power of ) satisfies: P mg1 (k+t i |k)+P mg2 (k+t i |k)+P UC (k+t i |k)+P B (k+t i |k)+P ag (k+t i |k)=P load (11) Among them, P mg1 (k+t i |k) represents the generator 1 at the current sampling time k to k+t. i Predicted power at time point, P mg2 (k+t i |k) represents the generator 2 at the current sampling time k to k+t. i Predicted power at time point, P UC (k+t i |k) represents the supercapacitor at the current sampling time k versus k+t. i Predicted power at time point, P B (k+t i |k) represents the lithium battery at the current sampling time k to k+t. i Predicted power at time point, P ag (k+t i |k) represents the auxiliary generator at the current sampling time k to k+t. i Predicted power at time point; P load This represents the load power.

6. The hierarchical control method for a high peak-to-average power ratio hybrid energy system of a multi-electric aircraft as described in claim 1, characterized in that, In step 3.2, the charging and discharging power constraint is: Among them, P mg1 (k+t i |k) represents the generator 1 at the current sampling time k to k+t. i Predicted power at time point, P mg2 (k+t i |k) represents the generator 2 at the current sampling time k to k+t. i Predicted power at time point, P UC (k+t i |k) represents the supercapacitor at the current sampling time k versus k+t. i Predicted power at time point, P B (k+t i |k) represents the lithium battery at the current sampling time k to k+t. i Predicted power at time point, P ag (k+t i |k) represents the auxiliary generator at the current sampling time k to k+t. i Predicted power at time point; P mg_MAX The maximum power of the main generator; P UC_MIN P UC_MAX These represent the minimum and maximum power of the supercapacitor, respectively; P B_MIN P B_MAX These are the minimum and maximum power of the lithium battery, respectively; P ag_MAX To supplement the generator power.

7. The hierarchical control method for a high peak-to-average power ratio hybrid energy system of a multi-electric aircraft as described in claim 1, characterized in that, In step 3.2, the charge constraint is: SOC UC_MIN SOC UC_MAX The minimum and maximum states of charge (SOC) of a supercapacitor. B_MIN SOC B_MAX The minimum and maximum states of charge (SOC) of a lithium battery. UC (k+t i |k) represents the value of k+t at the current sampling time k. i Predicted state of charge (SOC) at time t. B (k+t i |k) represents the generator 2 at the current sampling time k to k+t. i The predicted value of the state of charge at time t.

8. A hierarchical control method for a high peak-to-average power ratio hybrid energy system in a multi-electric aircraft as described in claim 1, characterized in that, The objective function of step 3.4 is: In the formula, k = 0, 1, 2…; N is the prediction step size; Q is the positive definite weighting coefficient matrix of the prediction output error; P mean P represents the system's average power, which is also the system's reference trajectory. mg1 (k+t i |k)P mg2 (k+t i |k) represents the current sampling time k against k+t i The predicted output values ​​of the two main generators at time 1.

Images